Dean Eckles (@eckles)

Data Visualization A Practical Introduction

Here’s a full draft of the upcoming second edition of my “Data Visualization: A Practical Introduction”: socviz.co

05.03.2026 22:54 👍 509 🔁 163 💬 12 📌 15

Ok I'm in a rabbit hole. If you search "how many decisions do we make in a day" the reported number is almost always 35,000, often reported that this is according to "multiple sources". Yet I can't actually find a single source that backs up that number. Anyone know where this number comes from?

05.03.2026 23:37 👍 18 🔁 5 💬 6 📌 0

Clinical trial reforms that once seemed radical How randomized controlled trials, preregistration, and results reporting became standard practice.

The idea of the need for a control group may have been radical at the time, but trust me it's now generally accepted. This is not some fringe believe by methodological hardliners; it's established practice.

www.clinicaltrialsabundance.blog/p/clinical-t...

05.03.2026 13:58 👍 17 🔁 3 💬 2 📌 0

Worthwhile new essay "Mathematicians in the Age of AI" by Jeremy Avigad, CMU professor and director of the NSF Institute for Computer-Aided Reasoning in Mathematics (ICARM) at CMU: www.andrew.cmu.edu/user/avigad/...

05.03.2026 13:55 👍 7 🔁 2 💬 0 📌 0

Just realized that with Abolish ICE at +11, it's polling over 30 percentage-points ahead of Trump approval. Guess that means if the media wants to be representative, it should give at least like 40% more airtime to people who support abolishing ICE to Trump supporters

04.03.2026 01:06 👍 1268 🔁 285 💬 12 📌 6

just realized that jupyter is probably dead as a concept. it's all md+scripts now.

04.03.2026 20:35 👍 81 🔁 8 💬 9 📌 7

Statistical Inference for Network Models A NetSci 2026 Satellite Symposium

Interested in giving a talk at @netsciconf.bsky.social but missed the main conference deadline? Submit your work to the Statistical Inference for Network Models (SINM) satellite! Send your one page abstract by March 15 and check out sinm.network for more details!

16.02.2026 04:50 👍 6 🔁 3 💬 0 📌 1

Looking forward to this!

04.03.2026 23:40 👍 5 🔁 2 💬 0 📌 0

Harsh Parikh will present work on how to transport estimated effects from one set of networks to another.

Shuangning Li will share new results on covariate adjustment in experiments.

So if you're working on networks+causality, consider participating. Abstracts due March 10th causnets.github.io

04.03.2026 22:05 👍 5 🔁 2 💬 0 📌 0

We've got 4 great invited speakers for our satellite event causnets.github.io at @netsciconf.bsky.social.
causnets.github.io/speakers/

In two nicely related talks, Christina Lee Yu & @vivianodavide.bsky.social will each present work on cluster-randomized designs in networks.

04.03.2026 19:56 👍 7 🔁 1 💬 2 📌 1

Is it primarily product differentiation? Did some econometrician in 1960 hear about Stein's theorem and go "we're having none of that; let's be the Amish"

Idgi, lexicographic preferences for unbiasedness over anything else is so alien.

29.11.2025 15:48 👍 5 🔁 1 💬 1 📌 0

If you're expecting there is something like a rigorous (or even just standard) diff-in-diff analysis in this paper, you may be disappointed...

03.03.2026 03:02 👍 2 🔁 0 💬 0 📌 0

Just a quick look suggests that:
1. Not a standard diff-in-diff plot, but the output some underreported regression "controlling" for GDP per capita.
2. Statistical inference is absent or neglects the country-level clustering.

03.03.2026 02:57 👍 5 🔁 0 💬 1 📌 0

I guess I'm more in the category of already having spent enough time reading other Twenge papers, so I have a very low prior on credibility

03.03.2026 02:52 👍 3 🔁 0 💬 1 📌 0

Robust empirical calibration of p‐values using observational data An official website of the United States government

In some settings, you may have some population of effects that are perhaps subject to the same distribution of biases. Then you could do something like this paper. There it is still critical to have cases where you know (or have a good estimate of) the true effect in some cases.

02.03.2026 17:43 👍 1 🔁 0 💬 0 📌 0

Health Recommendations and Selection in Health Behaviors (June 2020) - Consider a case in which a new research finding links a health behavior with good health outcomes. A possible consequence is take-up of this behavior among individuals who engage in othe...

My prior beliefs are that confounding is often large... So then it is helpful to have well identified studies in some particular area, which then might help us (if bias is small) be confident when doing observational research.

02.03.2026 17:41 👍 1 🔁 0 💬 1 📌 0

"If this event contributed anything, it simply made the ongoing death more obvious and less deniable for me personally. I consider the events of the last week a kind of death rattle of the old republic, the outward expression of a body that has thrown in the towel."

Don't skip the intro.

02.03.2026 16:48 👍 92 🔁 19 💬 1 📌 1

Mondrian Entered the Public Domain. The Estate Disagrees. The Mondrian Trust claims a 1930 painting is still protected—citing “dual copyrights,” Spanish law, and a misreading of the Copyright Act. Last February, I wrote about a troubling trend: estates and rightsholders making aggressive (and legally questionable) claims to works that have entered the public domain. From Tintin to Charlie Chaplin to Sherlock Holmes, the playbook is consistent: send threatening letters, cite convoluted legal theories, and expect recipients to back down rather than fight. A year later, the pattern continues. Now, it’s the work of Dutch abstract painter Piet Mondrian. And this time, the estate’s legal argument is more Dada than De Stijl. A Celebration Interrupted On January 1, 2026, Piet Mondrian’s iconic Composition II with Red, Blue, and Yellow (1930) entered the U.S. public domain, 95 years after its first publication. An art magazine marked the occasion with a Public Domain Day roundup, noting that the painting—along with works featuring Betty Boop, William Faulkner, and The Little Engine That Could—was now free to use, adapt, and celebrate. Standard Public Domain Day fare. The Mondrian/Holtzman Trust saw things differently. Piet Mondrian died in New York in 1944, leaving his estate to his close friend and fellow artist Harry Holtzman. The Trust is now administered by Holtzman’s descendants—professional stewards of Mondrian’s rights rather than heirs of the artist himself. When the art publication reached out to the Mondrian/Holtzman Trust to confirm the painting’s public domain status, the Trust responded that the work remains protected in the United States—and warned that reproducing it without permission constitutes copyright infringement. The correspondence, shared with me by Jennifer Jenkins, Director of Duke Law’s Center for the Study of the Public Domain, is a masterclass in legal confusion. Dual Copyrights and Spanish Detours The Trust’s argument rests primarily on the Uruguay Round Agreements Act (URAA), which restored U.S. copyright protection in 1996 to certain foreign works that had previously fallen into the public domain for failing to comply with U.S. formalities. According to the Trust: “Between 1944 and 1996 the work of Piet Mondrian not meeting the technical requirements of U.S. copyright law, was in public domain in the US. In 1996 the Uruguay Round Agreement Act… gave protection to the images by foreign artists which were first published outside of the United States between 1923 and 1977 (95 years from the date of first publication).” So far, so good. But then the letter takes a turn: “The duration of the U.S. protection for all other works… was for 70 years from the artist’s date of death. This is the reason of the dual copyright’s terms on the works of Piet Mondrian.” To be clear, copyright law isn’t a choose your own adventure book. There’s no such thing as a “dual copyright” regime under U.S. law—no parallel terms running simultaneously, no option to select whichever happens to be longer. Different categories of works receive different terms, but each work falls into one category based on its actual history. For works first published outside the United States between 1923 and 1977, URAA restoration provides the remainder of a 95-year term from first publication. For works created before January 1, 1978 that were neither published nor copyrighted, Section 303 of the Copyright Act supplies a term of life plus 70 years. The Trust appears to be blending these categories to suggest Mondrian’s works might enjoy whichever term is most favorable. But even on its own terms, the argument fails. Mondrian died in 1944. Any of his works subject to a life-plus-70 regime would have entered the public domain on January 1, 2015—more than a decade ago. Then the letter pivots to Spanish law, citing an 1879 statute and a decision of the Spanish Supreme Court (as one naturally does when disputing U.S. copyright over a Dutch painter’s work). Spanish law governs copyright in Spain. U.S. law governs copyright in the United States. And even under the Spanish rule the Trust invokes—life plus 80 years for certain authors—Mondrian’s works (he died in 1944) would have entered the public domain no later than 2025. What the Law Actually Says Under U.S. copyright law, works published before 1978 are protected for 95 years from the date of first publication. Composition II with Red, Blue, and Yellow was published in 1930. The painting entered the public domain on January 1, 2026—exactly when Duke’s Center for the Study of the Public Domain said it would. Yes, URAA restored copyright in Mondrian’s works in 1996 after they had originally lapsed into the U.S. public domain for failure to comply with formalities. But URAA restoration doesn’t extend the term. It fills the gap. Section 104A of the Copyright Act is clear: a restored work is protected only for “the remainder of the term of copyright that the work would have otherwise been granted in the United States.” For a work published in 1930, that remainder ran out on December 31, 2025. The Chilling Effect, Again What makes this episode especially frustrating is the target. The art publication wasn’t selling prints or producing merchandise. It ran an educational article celebrating works entering the public domain, complete with proper attribution and links to legal resources. For doing its homework and reaching out to confirm the painting’s copyright status, it received a warning letter. That outcome isn’t accidental. The Trust’s website invites anyone wishing to reproduce Mondrian’s work to “contact us about images you wish to reproduce… so that we can check on the copyright status and clear rights if needed.” The implication is clear: ask permission first, and the answer may depend less on the law than on who’s asking. Complexity is the strategy. If the rules sound uncertain enough, people keep asking. And if people keep asking, licensing fees keep flowing, even after the copyrights have expired. Confusion, after all, is far cheaper than litigation. The Mondrian Trust isn’t alone. From Tintin to Chaplin to Sherlock Holmes, estates keep claiming copyrights that have expired. Read the full story: “Their Copyrights Expired. The Legal Threats Keep Coming.” And Mondrian isn’t an isolated case. In recent weeks, similar post–Public Domain Day claims have surfaced from Fleischer Studios in connection with Betty Boop’s earliest appearances, and from representatives of Harold Lloyd’s estate concerning Safety Last! (1923), which entered the public domain over seven years ago. Different works, different theories—but the same objective: discourage use first, sort out the law later. A Clear Picture Mondrian’s abstractions are famous for their clarity—primary colors, clean black lines, precise right angles. The Trust’s legal arguments, on the other hand, are anything but. Composition II with Red, Blue, and Yellow is in the U.S. public domain. It has been since January 1, 2026. No amount of Spanish law or invented “dual copyright” theories changes that. If the Trust believes otherwise, it’s free to test that theory in court. For now, it appears content to keep its arguments abstract. As always, I’d love to hear your thoughts—leave a comment below, find me on social media @copyrightlately, or, if you’re asserting exclusive rights in a public-domain Mondrian, feel free to reach out directly to aaron@copyrightlately.com. You Might Also Like:

Mondrian Entered the Public Domain. The Estate Disagrees.

02.03.2026 16:46 👍 1 🔁 2 💬 0 📌 0

Part of the problem is that there are reasonably generic stories that are often true, and major problems. Like people try to do things that they think are good for them.

02.03.2026 14:29 👍 2 🔁 0 💬 1 📌 0

diagram of the Mechanical Turk showing how the operator could bend his body to avoid being seen as one cabinet door after the other was opened

Often seen in series of experimental studies, with improvements to validity brought out one at a time, but never all together. I liken this to the trick of Maelzel's Mechanical Turk: the human operator slid around inside the machine, never to be seen, as the showman opened one door after another.

17.02.2026 09:59 👍 9 🔁 2 💬 0 📌 0

Hegseth's "no stupid rules of engagement" (real quote) pairs well with Trump saying the US had some Khamenei successors in mind to back but the initial wave of American-Israeli strikes killed them.

02.03.2026 13:54 👍 222 🔁 54 💬 17 📌 2

Hmm I feel like people usually tell a story of why there is selection of a particular kind

02.03.2026 13:52 👍 1 🔁 0 💬 1 📌 0

time traveler from 12 months from now just sent me this

27.02.2026 21:25 👍 1620 🔁 205 💬 66 📌 34

You'll understand our campaign more from this launch video than anything else. We gave it our heart and soul. Give it a watch.

27.02.2026 13:37 👍 103 🔁 24 💬 5 📌 16

The power of descriptive norms...

27.02.2026 04:03 👍 32 🔁 4 💬 1 📌 0

Lost 19th century film by Méliès discovered at the Library | Timeless Library conservators recently made a startling discovery in a batch of decaying film reels -- a long-lost 1897 film by early cinema icon George Méliès. The French magician-turned-filmmaker's

Short form videos of absurd conflicts — like this 45 second film from 1897 about a man fighting with a robot — are frying young people's brains

blogs.loc.gov/loc/2026/02/...

27.02.2026 03:47 👍 4 🔁 0 💬 0 📌 0

Amazing article. Anyone who understands what "R is dense" means should read this. And even the others, because the article explains it well! 👏 💯

26.02.2026 15:14 👍 3 🔁 2 💬 1 📌 0

Per protocol analysis strikes again!

Folks, if you randomize but then don‘t analyze some of the people who got randomized (maybe because they didn’t adhere to instructions, maybe because they dropped out), randomization will no longer do all the heavy causal inference lifting.

25.02.2026 17:28 👍 190 🔁 56 💬 6 📌 3

It was a lot of fun building this tool! Something that can help faculty with a small but tedious step in preparing an NSF grant proposal: the conflict of interests spreadsheet. Drag-and-drop simplicity ftw 🚤

24.02.2026 22:00 👍 12 🔁 3 💬 0 📌 0

DoJ cases against protesters keep collapsing as officers’ lies are exposed in court String of embarrassing defeats for prosecutors as experts condemn DoJ effort to cast people as ‘violent perpetrators’

A MN protester accused of assaulting ICE officer had case dropped after vids showed no assault.

LA protester accused of assaulting ICE w/ “hat” had case dropped; judge said govt acted in “bad faith”

In Chicago, 92 ppl were arrested for assault/impeding ICE; 0 convicted

DOJ keeps lying and losing:

21.02.2026 21:22 👍 5889 🔁 2146 💬 117 📌 107

Dean Eckles

Latest posts by Dean Eckles @eckles